Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add OKHP3/skillz --skill auth-designgit clone --depth 1 https://github.com/OKHP3/skillzWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/okhp3/skillz/auth-design)<a href="https://agentmods.dev/skills/okhp3/skillz/auth-design"><img src="https://agentmods.dev/badge/skills/okhp3/skillz/auth-design/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/okhp3/skillz/auth-design"><img src="https://agentmods.dev/badge/skills/okhp3/skillz/auth-design.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00080 | $0.01244 |
| Opus 5 | $0.00040 | $0.00622 |
| Sonnet 5 | $0.00016 | $0.00249 |
| Haiku 4.5 | $0.00008 | $0.00124 |
Grade A, and why
auth-design scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Auth design
Two separate problems, constantly conflated, with different failure modes:
- Authentication: who is this? Mature, well-solved, and you should use a library
- Authorization: what may they do? Application-specific, and where nearly all real breaches originate
Teams spend most of their effort on the first because it is visible, and most of their vulnerabilities live in the second. Object-level authorization is the one to get right.
1. Do not build authentication from scratch
Password handling, session fixation, timing attacks, reset flows, and MFA are all solved and all easy to get subtly wrong.
Use the platform's framework, or an identity provider. If you are storing passwords yourself, use argon2id or bcrypt with the current cost parameters, never a general-purpose hash, never with a hand-rolled salt.
The flows people forget to secure, which are attacked precisely because they are afterthoughts: password reset, email change, account recovery, and MFA enrolment. A reset flow is a login bypass if the token is guessable, does not expire, or is reusable.
Done when: every path that can establish a session has been enumerated and secured.
2. Choose sessions or tokens deliberately
| Server sessions | JWTs | |
|---|---|---|
| Revocation | Immediate | Not possible before expiry |
| State | Server-side store | Stateless |
| Best for | Browser apps, anything needing instant logout | Short-lived service-to-service |
The revocation problem is the deciding factor and it is usually underweighted. A JWT valid for 24 hours is valid for 24 hours after you fire someone, after you detect a compromise, after the user clicks "log out everywhere". The usual workaround — a revocation list checked on every request — reintroduces the state JWTs were chosen to avoid.
Default to server sessions for browser applications. Use short-lived tokens with refresh where statelessness genuinely matters, and keep access token lifetime in minutes.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 116 lines · 80 tokens per session scan A d18ba388a20c
auth-design is a skill published in the GitHub repository OKHP3/skillz (3 stars, last pushed today), licensed MIT. It adds 80 tokens to every session and 1,244 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
feature-workflow
Phased workflow for delivering a Drupal feature or non-trivial change: plan, implement, review, test, finalize. Use when building a new custom module, adding a route/form/plugin, changing entity structure or schema, or any task touching more than a couple of files. Do NOT use for typo fixes, CSS tweaks, or single-line…
drupal-module-development
Use when creating or extending Drupal 11 custom modules, including scaffolding, service architecture, and dependency injection best practices.
scale-to-millions
Use this skill whenever the user is designing, reviewing, debugging, or answering interview questions about system/software architecture and scalability — e.g. "scale my app", "how do I handle more traffic", "design a URL shortener / news feed / chat system", "my API is slow under load", "should I shard my database"…
houtu-dependencies
Skill "houtu-dependencies" from lujiafa/houtu-project-skills, covering houtu framework — ai agent coding guide, core principles, code generation workflow (must be executed in order), step 1 — detect version & dependencies and step 2 — identify scenario & select module.
web-auth-integrator
Integrate authentication and authorization flows with provider-specific setup and RBAC safeguards.
web-api-tester
Test API endpoints for correctness, edge cases, auth handling, and OpenAPI contract compliance.